An Intelligent Cluster-Based Routing Scheme in 5G Flying Ad Hoc Networks

Applied Sciences

Mee Hong Ling¹  Muhammad Fahad Khan¹  Kok-Lim Alvin Yau¹  Yung-Wey Chong²  Muhammad Ali Imran³ 

[1] Department of Computing and Information Systems, School of Engineering and Technology, Sunway University, Petaling Jaya 47500, Malaysia;National Advanced IPv6 Centre, Universiti Sains Malaysia, USM, Gelugor 11800, Malaysia;School of Engineering, University of Glasgow, Glasgow G12 8QQ, UK;

关键词: flying ad hoc network;    deep Q-network;    reinforcement learning;    5G;    QoS;   

DOI  :  10.3390/app12073665

来源: DOAJ

【 摘 要 】

Flying ad hoc network (FANET) is an application of 5G access network, which consists of unmanned aerial vehicles or flying nodes with scarce resources and high mobility rates. This paper proposes a deep Q-network (DQN)-based vertical routing scheme to select routes with higher residual energy levels and lower mobility rates across network planes (i.e., macro-plane, pico-plane, and femto-plane), which has not been investigated in the literature. The main motivation behind this work is to address frequent link disconnections and network partitions in order to enhance network performance. The 5G access network has a central controller (CC) and distributed controllers (DCs) in different network planes. The proposed scheme is a hybrid approach that allows CC and DCs to exchange information among themselves, and handle global and local information, respectively. The proposed scheme is suitable for highly dynamic ad hoc FANETs, and it enables data communication between UAVs in various applications, such as monitoring and performing surveillance of borders, and targeted-based operations (e.g., object tracking). Vertical routing is performed over a clustered network, in which clusters are formed across different network planes to provide inter-plane and inter-cluster communications. This helps to offload data traffic across different network planes to enhance network lifetime. Compared to the traditional reinforcement learning approach, the proposed DQN-based vertical routing scheme has shown to increase network lifetime by up to $60\%$, reduce energy consumption by up to $20\%$, and reduce the rate of link breakages by up to $50\%$.

【 授权许可】

Unknown